The 2018 trade war and the end of dispute settlement as we knew it

Chad Bown 13 June 2019



Editor's note: This column first appeared as a chapter in the Vox eBook "Trade War: The Clash of Economic Systems Endangering Global Prosperity", available to download free here.

Future scholars may someday ask: Did the 2018 trade war cause the end of dispute settlement under the WTO? Or was it the failures of dispute settlement that caused the trade war?

For more than 20 years, the WTO’s dispute settlement system provided an orderly process for countries to resolve trade grievances and keep cooperation going.1 But in 2018, something broke down. 

In this column I explore why the inability to resolve underlying problems with the WTO itself deserves some of the blame. The main goal is to arrive at some potential lessons for a future system of dispute settlement.

The US deliberately pushed the WTO to the brink

Before turning to a critique of the WTO, I begin with the conventional wisdom. The US provoked a crisis in 2018 with three precisely targeted policy decisions that expertly poked holes in some of the WTO’s weakest spots. 

First, it imposed new tariffs – which it claimed would not be subject to international review – on nearly $50 billion of steel and aluminium imports. Formally, the US excused its new tariffs by triggering the WTO’s national security exception. The US administration has argued this exception is “self-judging” or “non-justiciable”, meaning that it cannot be questioned or benchmarked against externally verifiable economic evidence, unlike other opt-outs like antidumping or safeguards.2 But denying any outside check could lead to copycat behaviour and a protectionist spiral in which countries ignore even the most basic rules that limit tariffs. The result could be systemic failure.

Second, the US retaliated against another WTO member without first going through the formal dispute resolution process. Its tariffs on $250 billion of imports from China came after completing only an internal investigation. WTO rules require a country first win a dispute that requests the partner change its policies. The US could only be authorised to retaliate if China then refused to comply, and even then, the retaliation would be subject to WTO limits.

Third, the US initiated a procedure that could end the WTO’s system of resolving disputes. Countries currently have the right to appeal to the WTO’s standing Appellate Body (AB) if they disagree with a preliminary ruling. But the United States has refused to allow the appointment of new AB members as old members’ terms expire. By December 2019, the AB may not have enough members to issue rulings to appeals.3 But if no rulings are issuable, a forward-looking defendant country could simply trigger an appeal, put the legal case into permanent limbo, and eliminate the WTO’s ability to authorise tariff retaliation against countries that fail to comply.

Scholars have articulated the extraordinary economic and long-run institutional costs of these and other US policy actions taken in 2017-2018.4 Those costs are of first-order importance but will not be repeated here. 

Instead, the next sections explore the political-economic concerns with the WTO that may have contributed to these US actions.

China’s subsidies demanded US intervention of some form

The US imposed national security tariffs in part because of China’s state-driven economic model. In sectors like steel and aluminium, for example, China’s expansion increased from under 20% to over 50% of global production between 2002 and 2017. Yet, even as China’s domestic demand began to slow, production and its already formidable exports continued to increase. 

China’s subsidies and exports exacerbated three external concerns. Its potential global domination was worrisome on anti-competitiveness grounds because of its history of abusing international market power once acquired.5 Furthermore, US policymakers have become more sensitive to the fact that technology- and trade-induced shocks impose larger-than-expected adjustment costs on domestic communities and labour markets, and that the Chinese system may push ‘its share’ of those costs onto others (Autor et al. 2016).6 Finally, China got caught in US domestic politics. Steel and aluminium firms are geographically concentrated in American swing states, and US policymakers are historically responsive to their economic interests. And the industries’ older, mostly male workers may be part of the other recent US narrative over identity politics (Grossman and Helpman 2018).

US national security tariffs arose because others wouldn’t work or had been ruled illegal by the WTO

Other US policy options had been taken off the table for a combination of reasons.

The US had already emptied some of the WTO toolbox, but to little economic effect. Its use of antidumping tariffs had mostly stopped steel and aluminium imports directly entering from China. But China’s exports to third countries continued to rise – as did US imports from third countries – likely due to trade diversion and potentially trade deflection. 

But second, the US was unwilling to deploy a nondiscriminatory safeguard tariff – instead of a national security tariff – because earlier attempts had been thwarted by the WTO itself. The AB issued a series of legal rulings condemning US safeguards imposed over 1995-2003, including a 2002 US safeguard on steel.7

The US was also concerned a WTO dispute was too risky and potentially unwinnable

The US ruled out a formal dispute to stop Chinese subsidies, the first-best result, out of concern that the WTO was not well-equipped to constrain Chinese-style subsidisation.8 WTO subsidy disciplines can easily capture transparent, direct payments from a government agency to firms. But Chinese subsidies are different and often stem from a nuanced and complex combination of policies.

A recent OECD (2019) study of the downstream (finished) aluminium industry is illustrative. Its first key point is that primary aluminium is estimated to make up 75-86% of the cost of downstream products, and primary aluminium has benefited from highly subsidised Chinese coal. But second, China also imposed export restrictions on primary aluminium, implicitly subsidising Chinese downstream firms relative to their foreign competitors. China also rebated value-added taxes to exporters of downstream products without doing the same to primary producers.

The combined result was a heavily subsidised downstream, refined aluminium industry. But it is also one that the WTO legal system would have found challenging to address.9

The US imposed unilateral tariffs because that type of WTO dispute was also unwinnable

The idea that WTO dispute settlement was not well-positioned to tackle a suite of Chinese policies whose economic effect was to act against the spirit – if not the legal letter – of WTO rules applies similarly to the US unilateral tariffs on $250 billion of imports. 

The US economic argument was that China maintained high tariffs – for example, on automobiles – that contributed to foreign firms needing to access the Chinese market through foreign direct investment in lieu of exports. But investment, combined with China’s joint venture requirements and other regulatory barriers, created greater possibilities for the forcible transfer of foreign technology, industrial espionage, and theft of intellectual property (USTR 2018). Again, because it was a combination of Chinese policies – some, such as high-Chinese tariffs, that were not WTO-inconsistent when viewed in isolation – a WTO dispute seemed unable to solve the problem (see also Wu 2016).

However, one important counter-argument – and that the US  gave up on formal dispute settlement prematurely – is that it failed to appeal to the WTO’s non-violation nullification and impairment (NVNI) clause (Staiger and Sykes 2013). Loosely interpreted, NVNI is a legal provision by which the WTO could find China guilty of harming US trading interests even without having broken any explicit WTO rules. Nevertheless, there have been few attempts to use such an argument. With respect to China, it would also require that the AB grapple with economic models in order to assess how complex policy interactions taking place in a non-market economy adversely impact US economic activity. It is not as straightforward as asking the AB whether one Chinese policy crossed some red line.

It also would have been philosophically inconsistent for the US to ask the WTO to find another country guilty even though it hadn’t broken any explicit rules. The current US administration has argued repeatedly against any WTO attempts to encroach on national sovereignty.

The Appellate Body needs a reboot because it had gone astray

Much of the US action that could eliminate the Appellate Body was based on the concern that the WTO has engaged in judicial overreach. The argument is that the AB imposed obligations on the US and other countries that had never been agreed through 70 years of GATT or WTO negotiations.10

As noted, one example involved a series of rulings against US use of safeguards. In another backward-looking example, the US became frustrated that it lost dozens of WTO decisions over ‘zeroing’, or an approach to calculating antidumping tariffs. It has argued these rulings overly constrained the US ability to address unfair trade (Bown and Prusa 2011).

Other, forward-looking US worries involve China. In ongoing cases over China’s non-market economy status, the AB could further limit how the US deploys its antidumping tariffs. In other disputes, the AB may decide that Chinese state-owned firms are not a ‘public body’ and thus do not count as being a subsidy provider, limiting US use of countervailing duties.

The US view is that it was the AB’s own rulings that created the need for someone to step in. And the problems couldn’t be fixed without a crisis, because with the Doha Round of negotiations stuck, the WTO lacks a rules-making function to legislate corrections when the AB either makes a legal error or crosses a politically sensitive red line. 

To make its point, the US did two things. It imposed policies (national security tariffs and tariffs on China) that were even more problematic than the ones ruled illegal in the past. It also threatened the existence of the WTO dispute settlement system that had decided that those earlier policies were illegal.

Dispute settlement may be going back to the GATT

Dispute settlement was not always this legalistic. Under the General Agreement on Tariffs and Trade (GATT), the WTO’s predecessor from 1947-1994, disputes were typically resolved quite differently.

The key GATT distinction was that countries could veto the dispute at any stage. This included a veto from the defendant before the dispute began or right before adoption of a legal ruling. Because the resulting legal process was uncertain, disputes were addressed through negotiation or not at all. And it was also a system that benefited those with power. As the current US Trade Representative, Robert Lighthizer, has expressed an affinity for the GATT system (Lighthizer 2017), that may be where things are headed. 

Ironically it was the US that became increasingly frustrated with the GATT’s ineffectiveness in the 1980s. It turned to the “aggressively unilateral” Section 301 – the same law used to impose tariffs on China in 2018 – and demanded partners provide additional access for US exporters or face tariffs. Growing concerns with US unilateralism at the time helped lead to the Uruguay Round agreement and the creation of the WTO dispute settlement system.11 And over the ensuring 25 years, no countries could block the process of dispute resolution, and the WTO could authorise retaliation if countries were found not to comply with the rules. And critically important, it was rare for a dispute to ever reach the stage of retaliation before being resolved.

The other main problem with the GATT is that there was a lot more protectionism, much of it through “voluntary” export restraints (Hoekman and Kostecki 2009, Bown 2019). And while no major trade wars may have erupted, the GATT was never forced to confront today’s trade tensions between market-oriented and state-driven economies.

WTO dispute settlement is not gone yet, nor is the 2018 trade war. But if either goes, any evaluation of the effectiveness of the US strategy will also require a thorough assessment of the new dispute settlement system that replaces it. And that will require grappling with what the WTO system provided, warts and all.  

Author’s note: Thanks to Bernard Hoekman, Petros Mavroidis, Robert Staiger, and Mark Wu for comments. All remaining errors are my own.


Amiti, M, S J Redding and D Weinstein (2019), “The Impact of the 2018 Trade War on US Imports and Prices”, CEPR Discussion Paper 13564.

Autor, D H, D Dorn and G H Hanson (2016), “The China Shock: Learning from Labor Market Adjustment to Large Changes in Trade”, Annual Review of Economics 8(1): 205-240.

Bagwell, K, C P Bown and R W Staiger (2016), “Is the WTO Passé?”, Journal of Economic Literature 54(4): 1125-1231.

Bown, C P (2019), “US Special Protection in Historical Perspective: 1974-2018”, Peterson Institute for International Economics Working Paper, forthcoming.

Bown, C P and C Freund (2019), “Active Labor Market Policies: Lessons from Other Countries for the United States”, Peterson Institute for International Economics Working Paper 19-2, January.

Bown, C P and S Keynes (2019), “National Security and Trade – The Fear of Imitation”, Trade Talks, Episode 81, 18 April.

Bown, C P and M Kolb (2019), “Trump’s Trade War Timeline: An Up-to-Date Guide”, Peterson Institute for International Economics blog (last updated 24 February).  

Bown, C P and T J Prusa (2011), “US Antidumping: Much Ado About Zeroing”, in W J Martin and A Mattoo (eds), Unfinished Business? The WTO’s Doha Agenda, CEPR and World Bank, pp. 355-392.

Crowley, M A and J Hillman (2018), “Slamming the Door on Trade Policy Discretion? The WTO Appellate Body’s Ruling on Market Distortions and Production Costs in EU–Biodiesel (Argentina)”, World Trade Review 17(2): 195-213.

Dean, J M (1995), “Market Disruption and the Incidence of VERs under the MFA”, The Review of Economics and Statistics 77(2): 383-388.

Fajgelbaum, P D, P K Goldberg, P J Kennedy, and A K Khandelwal (2019), “The Return to Protectionism”, NBER Working Paper 25638.

Grossman, G M and E Helpman (2018), “Identity Politics and Trade Policy”, NBER Working Paper 25348.

Hoekman, B M and M M Kostecki (2009), The Political Economy of the World Trading System, Oxford University Press.

Horn, H, L Johannesson and P C Mavroidis (2011), “The WTO Dispute Settlement System 1995-2006: Some Descriptive Statistics”, Journal of World Trade 45(6): 1107-1138.

Lighthizer, R (2017), Remarks at CSIS Event on US Trade Policy Priorities, 18 September.

Maggi, G and R W Staiger (2011), “The Role of Dispute Settlement Procedures in International Trade Agreements”, Quarterly Journal of Economics 126: 475-515.

Mavroidis, P C (2016), “Dispute Settlement in the WTO: Mind over Matter”, in K Bagwell and R W Staiger (eds), The Handbook of Commercial Policy, Vol. 1A, Elsevier, North Holland.

OECD (2019), “Measuring distortions in international markets: the aluminium value chain”, OECD Trade Policy Paper No. 218.

Payosova, T, G C Hufbauer and J J Schott (2018), “The Dispute Settlement Crisis in the World Trade Organization: Causes and Cures”, Peterson Institute for International Economics Policy Brief 18-5.

Park, J-H (2016), “Enforcement and Dispute Settlement”, in K Bagwell and R W Staiger (eds), The Handbook of Commercial Policy, Vol. 1A, Elsevier, North Holland.

Pinchis-Paulsen, M (2019), “Trade Multilateralism and National Security: Antinomies in the History of the International Trade Organization”, NYU Working Paper, March.

Staiger, R W and A O Sykes (2013), “Non-Violations”, Journal of International Economic Law 16(4): 741–775.

Sykes, A O (2003), “The safeguards mess: a critique of WTO jurisprudence”, World Trade Review 2(3): 261-295.

USTR (2018), Findings of the Investigation into China’s Acts, Policies, and Practices Related to Technology Transfer, Intellectual Property, and Innovation Under Section 301 of the Trade Act of 1974, 22 March.

Wu, M (2016), “The ‘China, Inc.’ Challenge to Global Trade Governance”, Harvard International Law Journal 57(2): 261–24.


[1] Mavroidis (2016) explains the WTO dispute settlement process. Economic surveys of dispute settlement include Park (2016) as well as Bagwell et al. (2016: 1206-1218).

[2] For a discussion of the historical origins of the national security exception, see Pinchis-Paulsen (2019) and Bown and Keynes (2019).

[3] While the Trump administration has created a crisis by blocking all appointments, the Obama administration had already signalled discontent with the AB by blocking specific nominees and relying on some of the arguments below.   

[4] This includes the author elsewhere (see, for example, the collection in Bown and Kolb 2019). Fajgelbaum et al. (2019) and Amiti et al. (2019) provide model-based estimates of the economic costs of the 2018 tariffs. The Trump administration’s focus on trade deficits seems to drive its view that trade agreements have been unfair to the US. Finally, other costly US trade policies in 2017-2018 include withdrawal from the Trans-Pacific Partnership agreement, renegotiation of the Korea-US and North American Free Trade Agreements, and its potential new trade restrictions on automobiles.

[5] The US and other countries won two WTO disputes over China’s illegal export restrictions on raw materials and rare earth metals that had constrained foreign access to critical inputs used in advanced manufacturing.

[6] Of course, US spending on active labour market policies – a more efficient approach to facilitating adjustment – is also low relative to peer countries (Bown and Freund 2019).

[7] While the US imposed safeguards on solar panels and washing machines in 2018, indicating the constraint was nonbinding, the US government was forced to act because those cases were initiated by domestic industry. The last time the US government self-initiated a safeguard investigation was 2002, which resulted in the steel tariffs rebuked by the AB (Sykes 2003).

[8] The Obama administration filed a WTO dispute against China’s aluminium subsidies at the very end of 2016 that the Trump administration decided not to pursue. The US had pushed for a multilateral, OECD forum to address global (Chinese) steel overcapacity, but that also made little progress. A final argument against formal dispute settlement is the length of the process. Horn et al. (2011) find that, on average, three years elapse between the initiation of the dispute, the issuances of the panel and Appellate Body report, and the outcome.

[9] In a related dispute, Crowley and Hillman (2018) discuss how the AB found against the EU’s anti-subsidy tariffs despite Argentine export restrictions for soybeans resulting in a subsidy to Argentina’s downstream, processed soybean products.

[10] Payosova et al. (2018) describe these and other procedural concerns with the dispute settlement process. Maggi and Staiger (2011) examine the issue of judicial overreach by modelling WTO dispute settlement as potentially completing an incomplete contract.

[11] Mavroidis (2016) provides a discussion. 



Reginald Jones Senior Fellow, Peterson Institute for International Economics; CEPR Research Fellow


CEPR Policy Research